Coalbed methane (CBM) is the composite of components that may be adsorbed on coal at the naturally occurring conditions of reservoir pressure and temperature. As pressure is reduced, the CBM begins desorbing from the coal once the critical desorption pressure (CDP) is reached. CBM may consist largely of methane with smaller amounts of impurities, typically nitrogen and carbon dioxide and some minor amounts of intermediate hydrocarbons.
The capture and sale of CBM is a burgeoning industry both in the United States and internationally. In the CBM industry, a typical procedure for CBM recovery is often to penetrate the geologic formation with a substantially vertically drilled well and to either 1) case the hole, typically with steel casing through the coal interval followed by cementing the casing in place and perforating the interval all by methods commonly known in the petroleum industry, or 2) to case in a like manner the hole to the top of the coal and then drill through the coal, perhaps widening the hole drilled through the coal by a process known in the industry as underreaming. The former case is known as a cased completion and the latter is known as an open-hole completion. In either case, when producible water is present, typically water is pumped from the well through a tubing string to the surface in an attempt to lower the reservoir pressure, a generally necessary condition for releasing commercial quantities of CBM in most production scenarios. As reservoir pressure is lowered, a free gas phase will eventually form at the bottom of the hole and most of the free gas then will rise in the annulus between the casing and the tubing by gravitational forces, allowing the relatively buoyant gas to be produced at the surface from the annulus of the casing. The gas produced is then gathered and then typically sent to markets through pipelines.
Many CBM wells that will ultimately produce commercial quantities of coalbed methane do not do so when first put into production. The only gas produced initially in such wells is the relatively minute, generally noncommercial, quantity of gas that is in solution in the water at bottom-hole conditions of pressure and temperature. Most of this minute quantity will come out of solution as the produced formation water moves from conditions at the bottom of the hole to the lower pressure and typically different temperature at the surface. Such coal formations that do not produce gas initially beyond the amount contained in solution in the formation water are said to be undersaturated at reservoir conditions of pressure and temperature. Other definitions for undersaturated coals include: 1) when the storage capacity of the coal, typically expressed in standard (usually 14.7 psia and 60 deg F.) cubic feet of gas per ton of coal, exceeds the actual gas content of the coal expressed in the same units at reservoir pressure, or 2) when no free gas phase exists in the cleats and fracture system at reservoir conditions.
Storage capacity of the coal is typically determined in the laboratory from a captured sample of coal. A plot of the data is often made having the ordinate typically expressed in SCF/Ton and the abscissa being absolute pressure. This data is also often statistically fit with an equation to yield a curve, one such commonly used curve being known as the Langmuir isotherm as described in the reference of Yee et al., 1993. These “isotherms”, as the term implies, are measured at constant temperature generally corresponding to that of the reservoir from which the sample was obtained.
Unfortunately, some of the undersaturated CBM reservoirs may never produce commercial quantities of coalbed methane. One concern, therefore, is the determination of whether or not the coals in these undersaturated CBM reservoirs contain sufficient gas to be commercial. Such information, if it could be determined expediently on a given well in an exploratory area, could prevent the drilling of a large number of wells in the specific area that may never produce economic quantities of CBM. As mentioned above, one common method of making that determination is through the process of obtaining a sample of the coal itself, perhaps by coring the coal, and subsequent detailed measurement of gas content of that sample in a laboratory or otherwise. This technique is typically expensive, and can require specialized drilling equipment and personnel. Additional expense may be incurred when the core samples are sent to commercial or private laboratories for analysis. The results of such core analyses are not immediately available, sometimes taking months of desorption time. Also, because core analysis may be too expensive for a large amount of sampling to be taken from a particular well, samples, hoped to be representative, are often selected. Consequently, there is the potential problem of the core samples not being representative of the formation even nearby the well from which the core was cut; and there is an additional problem of how representative the samples will be of the formation at some distance from the well. The CBM industry is replete with examples of how gas content can drastically change over relatively short distances. It is typically neither economically practical nor timely to have every well cored and analyzed.
The results from a sample of the coal itself, perhaps from the coring process, can also be very inconsistent from what is ultimately observed during production. During a coring or other sampling operation, not only are samples of coal pulled for determining gas content in the laboratory, but also a specific sample or a composite sample, possibly made up from drill cuttings, may be gathered and this sample used to determine storage capacity of the coal. This can involve tedious and expensive laboratory processes. The commercial or private laboratory may then compare the gas content measured in some samples with the storage capacity determined from another sample and estimate the degree of saturation of the coal. As explained above, if the measured gas content is less than the storage capacity, the coal is said to be undersaturated with gas, and the laboratory will typically determine the pressure at which the gas content intersects a plot of the storage capacity data. The resulting pressure is typically referred to as the critical desorption pressure (CDP). The CDP is the reservoir pressure at which CBM will start to desorb from the coal with reduction of reservoir pressure, become a gaseous phase, and begin to become capable of production in commercial quantities.
Unfortunately, the value of CDP determined by the laboratories, too frequently, has been grossly in error from what was ultimately observed when the wells were produced. The present inventor has identified such error in the coring and subsequent laboratory analyses of several of approximately ten wells, analyzed under traditional core analysis using different laboratories. Some analyses have indicated that the reservoirs are saturated at reservoir pressure, yet these reservoirs have not produced any commercial quantities of gas until the reservoir pressure has been drawn down to at least 50 to 60% of the initial reservoir pressure before reaching the CDP. Some of the analyses indicate that the gas contents exceed the storage capacities of the coals at reservoir pressure, something that appears to defy an adequate physical explanation.
In summary, coal sampling, coring, and subsequent core analyses as described above may lead to results that are not only time consuming and expensive to obtain, but also they can be highly questionable and frequently inconsistent when used for individualized analysis. For individualized analysis, due to uncertainty, the better use for coal sampling, coring, and core analyses may not come from individual assessments but instead from multiple assessments from which composite isotherms are constructed for a given geological region by averaging of the data and statistically demonstrating the uncertainty. This has been done in the Powder River Basin (PRB) by the United States Bureau of Land Management (BLM) as described in the reference to Crockett and Meyer, 2001. For example, from some 40 samples, the BLM has constructed an averaged synthesized isotherm for samples measured in the PRB representing these 40 samples. Even from such a relatively large number of samples, and ignoring the cost challenges to achieve such data, this effort highlights the challenges in a coal sampling approach because uncertainty in the data still exists. In fact this data shows significantly differing isotherms that represent one standard deviation on either side of the mean curve.
Another problem under traditional analysis can, and does, occur in some undersaturated CBM reservoirs when one tries to demonstrate, perhaps through individual testing or small-scale pilots of several adjacent wells, that the well(s) will ultimately produce commercial quantities of CBM. A long and uncertain dewatering period, even under the best of circumstances, may be required before any commercial quantities of CBM are produced. This can lead to long periods of evaluation time. In some areas where there is high permeability and strong aquifer support, such as can be the case in the PRB, one well cannot draw down the pressure sufficiently to ever reach the CDP in any sort of practical or economic time frame. In response to this problem and in an effort to evaluate their leases, most operators have drilled costly (multi-million dollar) multiple-well pilots in an effort to cause interference between wells so that these wells, in combination, can draw the pressure down sufficiently to reach the CDP by exceeding the water influx into the pilot area. Some of these pilots have been successful in the PRB, but some of the pilots have been dewatering for over three years without yet producing commercial quantities of CBM. This dewatering is done at considerable cost of equipment and power to pump wells, at a financial cost of deferred revenues and with the uncertainty that the ultimate resource to be found may not be sufficient to be profitable.
The practical challenges of laboratory involvement and sampling difficulty known to exist in a coal sampling-based technique are perhaps highlighted by reference to U.S. Pat. No. 5,785,131 to Gray. Although this reference involves techniques for sensing formation fluids as in gas-oil systems when the fluid itself is of interest, as it relates to the very different aspect of sampling solids containing a substance of interest, it proposes a system for pressurized capture of the samples from entrained particles during drilling. In the reference, these particles of coal or the like are captured and tested on site to avoid some of the mentioned challenges of laboratory testing. As it relates to the solids such as are of interest in the present invention, however, this reference still relies on a capture of the entrained particles and as such it is subject to the uncertainties and other practical limitations discussed above.
Another alternative to those techniques based on sampling of the coal itself involves the use of mudlogging during drilling to obtain, at least a qualitative indication of the presence of CBM. Some have even tried to quantify results (Donovan, 2001), but these techniques can leave much to be desired and problems can exist because the system is not usually closed, thus allowing unmeasured gas to escape. Gas-free drilling water is also typically mixed with formation water of different gas content. Further, particle size can need to be estimated, drilling speed recorded, etc. Then, too, results observed by the inventor for the PRB seem to indicate gas contents that are typically far in excess of those observed. Finally, such techniques provide, at best, an estimate for gas content of the coal and do not provide the practical accuracies desired, neither do these techniques provide an estimate for CDP.
Other than the coal sampling-based techniques mentioned above, efforts (e.g., see Koenig, 1988) have included attempts to determine CDP by producing the well and dropping the pressure, perhaps by bailing or by a pump lowered into the well until gas starts being produced. These techniques can be fraught with problems, some of which are: 1) if a pump is used in the well, its capacity may not be sufficient to draw the well down in a practical testing time frame to determine when gas starts being produced; 2) as the liquid level drops in the well, air may be pulled into the casing from the surface, if the casing is open at the surface, because the pressure in the casing will likely be lower than the atmospheric pressure at the surface, or if the casing is isolated from atmospheric pressure (e.g., shut in) a vacuum may be drawn on the well and a negative gauge pressure (in this document gauge pressure will refer to measurement of pressure above atmospheric pressure where zero gauge pressure would correspond to atmospheric pressure) may result until there is sufficient release of gas from the coal to overcome the vacuum being drawn by the falling liquid level; and 3) by the time the pressure is drawn down sufficiently to see gas production at the surface, the reservoir may already be affected by two-phase flow that may lead to complications in interpretation. This can also produce results inconsistent with later production history.